Cluster Method Using A Combination of Cluster K-Prototype Algorithm and Genetic Algorithm for Mixed Data

Abstract

Clustering is one of the main methods in data mining that useful to explore the data. One conventional clustering methods namely the K -Means algorithm efficient for large dataset and numeric data types but not for categorical data type. K-prototype algorithm eliminates the limitations of the numerical data but can also be used on categorical data. But the solutions generated by the algorithm is a local optimal solution in which one of the causes is the determination of the initial cluster’s center. Deal with these problems, the genetic algorithm was proposed for solving this global optimasitation problem. The results of the study indicate that the cluster’s center optimization with genetic algorithm success to improve the accuracy of the results of the cluster with K–Prototype algorithm.